Artificial Inteligence for Lyrics Comprehension
The aim of LyrAIcs is to build an AI based recommendation engine for song lyrics that will enrich the potential of existing music and song classifications with the content extracted from the lyrics text, using Artificial Intelligence and Natural Language Processing technologies. It will be built combining the algorithms generated by the POSTDATA-ERC funded project, devoted to automated poetry analysis and classification.
The algorithms generated by POSTDATA (Grant Agreement: 679528) are able to analyze and classify Spanish poetry in three levels: 1) metrical and rhythmic 2) semantic and conceptual and 3) emotional and sentimental. These algorithms have been built combining existing open-source libraries, together with the latest Deep Learning libraries and they are able to extract poetry features classified following a poetry ontology that can be used for processing the texts of the lyrics and enrich the RS with qualitative and real-time analyzed metadata.
LyrAIcs proposes to create a product that will transform all these powerful algorithms into a webservice that will be commercialized into the music market, by linking its APIs to the main Music Streaming Services. The automated generated metadata extracted from the lyrics analysis will enrich their existing Recommendation Systems real time and without the need of manual tagging. Algorithms will be able to “learn” and improve efficiency and accuracy by being trained with the existing metadata and tags and by the interaction with new datasets, classification and continuous interaction with data provided by users.
Our engine will target a market of +1,5Bn songs/year, as it will be mainly focused on the Spanish content of the streaming world. No other tool or platform offers content and language analysis for Spanish (whereas in English there are already beta solutions available), and it is the highest growing market (16%) with the Latino music and the number of consumers listening to that music increasing exponentially.
This project has received funding from the European Research Council (ERC) under the European Union’s
Horizon 2020 research and innovation programme (grant agreement No 964009).
Contact us for more information.